Stands up product analytics from decision questions backward - a north-star metric with an input-metric tree, an object-action event taxonomy with a governed tracking plan, and the priority analyses to build first: activation funnels, cohort retention curves, and feature adoption correlated with retention. Use when someone asks "what events should we track", "set up our analytics taxonomy", "where do users drop off before first value", or "what's our activation moment". Do NOT use for growth-accounting decompositions of MAU changes - use growth-accounting instead - or for deep funnel diagnosis on existing data - use funnel-analysis instead.
Click to play with sound.
---
name: Product Analytics
description: Stands up product analytics from decision questions backward - a north-star metric with an input-metric tree, an object-action event taxonomy with a governed tracking plan, and the priority analyses to build first: activation funnels, cohort retention curves, and feature adoption correlated with retention. Use when someone asks "what events should we track", "set up our analytics taxonomy", "where do users drop off before first value", or "what's our activation moment". Do NOT use for growth-accounting decompositions of MAU changes - use growth-accounting instead - or for deep funnel diagnosis on existing data - use funnel-analysis instead.
---
# Product Analytics
You cannot improve what you do not measure, but most teams measure the wrong
things badly. A messy event taxonomy is technical debt that misleads every
decision built on it. This skill sets up analytics that actually inform product
work.
## Start With Questions, Not Events
Before instrumenting anything, write the decisions analytics must inform:
- Where do users drop off before first value?
- Which features correlate with retention?
- What does an activated user do that a churned one did not?
Instrument backward from these questions. Tracking everything "just in case"
produces noise nobody queries.
## Event Taxonomy
… install to load the full skill